helping mobile operators make the best use of network resources
MODERNIZING TELCO BUSINESS FOR 5G
To help mobile operators make the best use of network resources, Mavenir developed the Open RAN Intelligent Controller (RIC), uniquely designed with deep knowledge of the Radio Access Network (RAN) domain, Artificial Intelligence and Machine Learning (AI/ML), and cloud-native, software-defined networking.
Designed based on O-RAN standards, the RIC is a new function that controls Radio Resource Management (RRM) decisions for the RAN. The RIC allows for RRM functions to be migrated from traditional vendor-proprietary hardware, and can be deployed which extensible applications alongside the platform functions to provide tangible business results for the operator.
The Mavenir RIC adds strategic value and differentiation to the operator network by providing a framework that automates RAN operational workflows while also optimizing end-to-end network performance.
The RIC offers the following benefits for the operator:
- Provides open APIs for RIC application development, enabling proactive network resource management, and allowing fine-grained UE policy deployments
- Enables closed-loop end-to-end network automation with standards-based converged network analytics
- Enhances network management and services orchestration with the ability to scale up and down through cloud-native, distributed architecture
- Improves Quality of Service (QoS) with near real-time and granular RAN control
THE POWER OF NETWORK INTELLIGENCE AND AUTOMATION
The RIC, together with the apps, provide for a smarter network that can:
- Lower TCO by leveraging intelligence to simplify operational workflows and improve radio spectrum efficiency
- Simplify deployments with proven, end-to-end domain knowledge for use case realization
- Enable 5G revenue-generating services by providing granular and programmable control over the 5G mobile network
- Optimize the user experience by applying Intelligence to increase throughput, decrease latency, and extend coverage to individual subscribers
MAVENIR’S RAN INTELLIGENT CONTROLLER (RIC) AND APPLICATIONS
Mavenir provides both RIC types (Near-Real Time and Non-Real Time) as defined in O-RAN-defined architecture.
Near-RT RIC
Mavenir’s near-RT RIC is the first to control RAN activity at BOTH the cell and individual user level. The containerized application hosts trained AI/ML applications to infer and control O-RAN elements in near-real-time.
The Near-RT RIC is responsible for fine-grained RRM of control-plane and user-plane of the RAN protocol stack at a per-UE level over the E2 interface. The Near-RT RIC is typically deployed at the edge of the RAN and controls RRM decisions for the RAN functions via xApps at near-real-time granularities, typically ranging from 10 milliseconds to 1 second.
Non-RT RIC
Mavenir’s non-RT RIC is a containerized application that uses advanced machine learning algorithms to optimize network performance and train ML models using long-term RAN data for dynamic and adaptive policy and control.
The Non-RT RIC is responsible for setting high-level declarative policies and intents, sending configuration recommendations, and use-case-specific prediction/enrichment information via rApps to the Near-RT RIC over the A1 interface. The Non-RT RIC is hosted in a Service Management and Orchestration Framework (SMO), typically deployed in a centralized cloud, which is responsible for RAN FCAPS operations and orchestration of platform infrastructure resources.
Applications
The RIC opens the door to a rich set of applications operationalized via an “app store,” including rApps to optimize network performance and xApps to infer and control O-RAN functional elements.
rApps include offline ML model training based on data collected in the Service Management Orchestration (SMO) and feedback received from the near-real-time RIC. All C-SON functionalities are provided as rApps in the non-RT RIC.
xApps allow the near-RT RIC to optimize the radio resource management decisions for control-plane and user-plane functionalities across the layers of the RAN protocol stack on a per-user level. Each xApp offers radio resource management solutions to optimize specific RAN functionalities using data-powered AI and analytics tools and the incorporation of machine learning.
Putting the Pieces Together
END-TO-END DOMAIN EXPERTISE
Automation + Openness + Intelligence = Real Business Value
Mavenir brings several domains together to modernize the CSP network and deliver real business value.
Telco Cloud (Automation)
Telco Cloud Automation is a must-have for 5G. Mavenir is a pioneer in cloud-native, containerized deployments at massive scale. Mavenir’s RIC is part of a full, cloud-native stack that includes Service Management Orchestration, xApps and rApps, a Webscale Platform (for Caas, Paas, and Telco layer Integration. This end-to-end functionality is something that takes years of experience to build and get right.
Radio Access Networks (Open Networks)
Mavenir’s award-winning Open vRAN solution brings increased business agility with network elasticity and flexibility in radio access networks with the world’s first fully containerized, virtualized Open RAN Split 7.2 architecture. Eliminate vendor lock-in and leverage open interfaces, virtualization, and web-scale containerization to support various deployment scenarios – including Public Cloud, Private Cloud, resulting in a 37% savings in TCO over 5 years. (Source: Senza Fili 2021)
AI/Machine Learning (Intelligence)
CSPs can make the best use of the network by combining deep Telco domain expertise with finely-tuned Artificial intelligence (AI) and Machine Learning (ML) algorithms. The RIC uses various ML models such as anomaly detection, time series prediction, clustering, and Bayesian optimization to enable numerous RAN control use cases.
Intelligent Control Expertise
- Integration of RIC with RAN
- Development and integration of Applications (use cases)
- Development and tuning of ML Algorithms to maximize use case results
RAN Domain Expertise
- CU/DU integration with RIC
- Extract and analyze RAN measurements/data
- Configuration of RAN nodes completing control loop
Telco Cloud Expertise
- Data collection/extraction from OBF
- Integration to CI/CD LCM and Configuration to complete control loop
Mavenir Videos
RAN INTELLIGENT CONTROLLER (RIC)
Mavenir demonstrates an O-RAN compliant AI-powered closed-loop Near-RT RIC platform along with the traffic steering that substantially reduces network overhead and improves cell capacity savings. The demo validates a real-time learning framework using cloud-native network APIs for xApp to xApp communication and integration of 3rd party apps.
Non-RT RAN Intelligent Controller
In trials of the Mavenir non-RT RIC, a Tier 1 mobile network achieved significant improvements in network performance, which results in enhanced user experience.
Comparing one week of data with and without the RIC, the reference KPIs versus the KPIs optimized with the RIC showed:
- A 20% increase in user downlink throughput for similar call volumes.
- An increase in total user payload (measured in GB) by 5.3 percent.
- A CQI distribution increase of 3.12%
- MCS improvement was up 2.85 percent (DL) and increased 5.9 percent (UL)
- PRB utilization was reduced 13.6 percent (DL) and lowered by 1.4% (UL)
Near-RT RAN Intelligent Controller
In the O-RAN India Plugfest 2021 hosted by Airtel, Mavenir showed a live demonstration of the world’s first O-RAN standards-compliant Near Real-Time RAN Intelligent Controller (Near-RT RIC) with an AI-powered extensible application (xApp). This xApp controls the traffic steering functionality of a 5G Radio Access Network (RAN), a key feature that is responsible for managing the connectivity and mobility of users in the network. The measured results showed the following compared against SON-based RAN handover algorithms:
- Improved mobility overhead KPI by reducing the number of handovers by around 50%
- An increase in the throughput KPI by over 20% for cell-edge UEs
BUILDING THE WORLD’S FIRST O-RAN-COMPLIANT, AI-POWERED, CLOSED-LOOP NEAR-RT RIC
How Mavenir drastically improved RAN performance by reducing network overhead by more than 50% while increasing cell-edge capacity and user throughput by over 20%.
- Anomaly Detection
- Event Correlation
- Dynamic Policy Tuning
- Per UE Control
- Parameter Tuning
- Steering and Load Balancing
- Prediction Engine
- Auto RCA & Recommendation
- Auto SLA Management
Press releases
News Articles
Blogs
On-Demand Resources
